Typical operation of a continuous jet printer may be described as follows: electrically conductive ink is kept under pressure in an ink reservoir. The ink reservoir feeds a chamber that contains ink to be stimulated by means of an ink stimulation device. Working from the inside outwards, the stimulation chamber comprises at least one ink passage to a calibrated nozzle drilled in a nozzle plate: pressurised ink flows through the nozzle, thus forming an ink jet.
The ink jet thus formed breaks up at a well defined point downstream the nozzle plate and produces ink droplets at regular time intervals under the action of the periodic stimulation device housed in the ink chamber; this forced fragmentation of the ink jet is induced at a point called the drop break up point by the periodic vibrations of the stimulation device located in the ink contained in the ink reservoir.
Starting from the break up point, the continuous jet is transformed into a sequence of ink drops. A variety of means is then used to select drops that will be directed towards a substrate to be printed or towards a recuperation device commonly called a gutter. Therefore the same continuous jet is used for printing or for not printing the substrate in order to make the required printed patterns.
Such continuous jet printers may comprise several print nozzles operating simultaneously and in parallel, in order to increase the print surface area and therefore the print speed.
Usual drop selection means comprise a first group of electrodes close to the break up point called charging electrodes, the function of which is to selectively transfer a predetermined electrical charge to each drop. All drops in the jet, some of which having been charged, then pass through a second arrangement of electrodes called the deflection electrodes generating an electrical field that will modify the trajectory of the drops depending on their charge.
This electrostatic deflection of liquid drops issued from fragmentation of a continuous jet is a solution widely used in ink jet printing. For example, the deviated continuous jet variant described in document U.S. Pat. No. 3,596,275 (Sweet) consists of providing a multitude of voltages to charge drops with a predetermined charge, at an application instant synchronised with the generation of drops so as to accurately control a multitude of drop trajectories. The positioning of droplets on only two preferred-trajectories associated with two charge levels results in a binary continuous jet print technology described in document U.S. Pat. No. 3,373,437 (Sweet).
For all these devices, the charging signal is determined according to the trajectory to be followed by the drop, and other factors. The main disadvantages of this concept for use with multiple jets are firstly the need to place different electrodes close to each jet, and secondly to control each electrode individually.
Another approach consists of setting the charging potential and varying the stimulation signal to move the jet break up location: the quantity of charge carried by each drop and consequently the drop trajectory will be different, depending on whether the drop is formed close to or far from a charging electrode common to the entire array of jets. The set of charging electrodes may be more or less complex: a multitude of configurations is explored in document U.S. Pat. No. 4,346,387 (Hertz). The major advantage of this approach is the mechanical simplicity of the electrode block, but transitions between two deflection levels cannot be easily managed: the transition from one break up point to another produces a series of drops with uncontrolled intermediate trajectories.
Solutions have been considered to overcome this difficulty comprising a modulation of the break length in EP 0 949 077 (Imaje), but with a tight tolerance on the break up length (typically a few tens of microns) that is difficult to control; or management of partially charged portions of the jet with a length equivalent to the distance separating two clearly defined break up locations in EP 1 092 542 (Imaje), but this requires management of two break up points and the useful drop generation frequency has to be reduced, with the production of unusable jet segments.
In general, even for recent developments such as developments made by the Kodak company for its drop generator based on a thermal stimulation technique allowing exceptional drop production ways (for example EP 0 911 167), the solutions put forward always have the problem of transitions between the deflected position of the jet and the undeflected one.
One alternative suggested the presence of different sized drops and selective deflection according to the drop sizes by crosswise projection of an airflow, as described in US 2003/0222950. However in this case, the production, circulation and recovery of a uniform airflow are difficult to implement without increasing air induced fluctuations along the trajectory of the drops.